This proposal seeks to extend the ability of fMRI to accurately and quantitatively image high-resolution spatial structure in the human brain, and, at the same time, to expand mathematical and computational understanding of the basic functional architecture of visual cortex. Pilot data indicates that there is a super-modular structure linking multiple topographic maps in macaque and human visual cortex. This V1-V2-V3 complex can be modeled as a single map structure, and it is suggested that the MT-MST complex may be a second instance of the same super-modular organization. In order to test and extend this modeling, the new, state-of-the-art 7T fMRI scanner at Harvard MGH will be used, along with the existing 3T facility, to produce high resolution, minimal distortion, wide visual field reconstructions of human V1, V2, and V3 topographic mappings. A variety of specialized methods is proposed to improve the signal-to-noise ratio and to minimize spatial distortion and other artifacts which exist in current fMRI imagery. Special head coils designed to optimize recording from human occipital cortex have been prototyped. A novel, wide-field visual stimulation device has been designed, which will allow, for the first time, stimulation of the entire visual field from fovea to far periphery within the narrow bore of the fMRI scanner, rather than the para-foveal only stimulation that has been performed to date. All software and distortion correction methods developed under this proposal will be made available and maintained on the MGH web site, as well as source code for modeling the V1-V2-V3 complex, and all instrumental and electronic details for fMRI resolution enhancement and head coil design, as well as the novel full field visual stimulator designed for fMRI use will be made publicly available. The goal of this proposal is to join the efforts of the leading research groups in topographic map modeling and fMRI imaging, in order to extend and validate current understanding of the spatial structure of primary and extra-striate visual cortex and to extend the precision and resolution of fMRI imaging--two goals which can only be accomplished by a joint modeling and experimental collaboration. [unreadable] [unreadable]